Upload Data

Expression Data

An unlabelled protein expression data file called “pro_edata.csv” was uploaded to pmart. The protein data was not rolled-up from peptide data. The column that designates unique molecules was marked as “Reference”. The original scale of the data was log2 and remained unchanged. The value to denote missing data was “NA” and the expression data was not already normalized.

Biomolecule Information

No biomolecule information was uploaded.

Sample Information

An associated sample information file was also uploaded called “pro_fdata.csv”. Trimmed sample names were not used. The column in the sample information file which indicates sample names was designated as “SampleID”.

Group Samples

Grouping Information

This table summarizes all the user specified main effects or covariates in pmart. The “Selected Column Name” denotes the name of the column in the sample information file assigned as a main effect or covariate.

Main Effect or Covariate Selected Column Name
First Main Effect Condition
Second Main Effect None selected
First Covariate None selected
Second Covariate None selected

Number of Samples per Main Effects

Data Summary

Summary Table

The first column in the table below denotes a property of the peptide data, and the “Data” column states that property’s value.

Data
Class proData
Unique SampleIDs (f_data) 10
Unique References (e_data) 2578
Rows (e_meta) NA
Missing Observations 3434
Proportion Missing 0.133
Samples per group: Mock 3
Samples per group: Infection 7

Missing Value Table

In the table below, the first column denotes the sample and the second is the missing number of observations. The third column represents the second as a percentage of the total number of observations for that sample.

Missing Observations Proportion Missing
Mock1 194 0.075
Mock2 203 0.079
Mock3 171 0.066
Infection1 271 0.105
Infection2 281 0.109
Infection3 416 0.161
Infection5 316 0.123
Infection6 379 0.147
Infection7 442 0.171
Infection9 761 0.295

Filters

Filters were applied. A total of 5 filters were applied. See the table below for a descriptions of the filters and the order.

Summary of Applied Filters

Order Filter Type Parameters Summary
1 Molecule Filter Biomolecule Min Number Molecules: 2 0 biomolecule(s) were filtered.
2 CV Filter Biomolecule Max CV: 120 2 biomolecule(s) were filtered.
3 imd-ANOVA Filter Biomolecule Min ANOVA: 2 & Min G-Test: 3 151 biomolecule(s) were filtered.
4 rMD Filter Sample P-Value Threshold: 0.001 & Metrics Used: MAD, Kurtosis, Skewness, Correlation, Proportion_Missing 0 sample(s) were filtered.
5 Custom Filter Sample or Biomolecule 1 sample(s) were filtered. 0 biomolecule(s) were filtered. 0 protein(s) were filtered.

Molecule Filter

A molecule filter was applied to the data, which removes biomolecule(s) (References) not having at least the minimum number of samples (Min Number Molecules).

CV Filter

A coefficient of variation (CV) filter was applied to the data which removes biomolecule(s) (References) with a CV greater than the threshold (Max CV).

imd-ANOVA Filter

An ANOVA filter can be applied to the data which removes biomolecule(s) (References) not having at least a minimum number of non-missing values per group (Min ANOVA). Additionaly, an IMD (independence of missing data) filter can be applied to the data, removing biomolecules not having at least a minimum number of non-missing values (Min G-Test) in at least one of the groups.

Proteomics Filter

Proteomics filters were not applied to the data.

rMD Filter

A robust Mahalanobis distance (rMD) filter was applied to the data, removing sample(s) (SampleIDs) with an associated rMD-associated p-value less than the threshold (P-Value Threshold). Metrics used to calculate the p-value are also included (Metrics Used).

Custom Filter

Custom filters can be used to remove samples, biomolecules, or proteins.

Normalization

Peptide data was normalized.

SPANS

SPANS was run to determine optimum normalization parameters.

Manual

Manual normalization was used to normalize the data.

Attribute Value
Subset Function all
Subset Parameters
Normalization Function mean

Protein Roll-Up

Protein Roll-Up was not conducted.

Statistical Analysis

The statistical analysis step was run.

The following groups were compared: Mock_vs_Infection. Reported below are the parameters used in the statistical analysis, followed by the number of siginificant biomolecules for each comparison.

Attribute Value
Test Method anova
Multiple Comparison Adjustment holm
Significance Threshold 0.05
Comparison Up_total Down_total Up_anova Down_anova Up_gtest Down_gtest
Mock_vs_Infection Mock_vs_Infection 718 328 718 328 0 0